Taurolidine and oxidative stress: a rationale for local treatment of mesothelioma

Cell cultures

We used two human mesothelial non-neoplastic cell lines (HMC and MET5A) and two established human malignant mesothelioma cell lines (MMB and MMP). As controls we used human dermal fibroblast (HDF) and human lung carcinoma (A549). Primary HMCs were obtained from patients with congestive heart failure and cultured in Ham's F12 medium (Sigma-Aldrich, St. Louis, MO, USA) supplemented with 10% fetal bovine serum (FBS; Life Technologies, Rockville, MD, USA). MET5A were purchased from the American Type Culture Collection (ATCC; Manassas, VA, USA), and cultured in Medium 199 (Sigma-Aldrich) supplemented with 10% FBS. MMB and MMP cells were derived from pleural effusions of malignant mesothelioma patients and cultured in Ham's F12 medium supplemented with 10% FBS. HDF were obtained from a healthy donor and cultured in Ham's F12 medium supplemented with 10% FBS. A549 were purchased from ATCC and cultured in Ham's F12 medium supplemented with 10% FBS. Cells were grown at 37°C in a 5% CO₂-humified atmosphere. MyrAkt-MMB were obtained by transfecting MMB cells with Addgene plasmid 9008 (pcDNA3 myr-HA-Akt1), using lipofectamine 2000. Transfectants were selected by G418 for 3 weeks.

Chemicals

Glutathione mono-ethylester (GSH), N-acetyl-l-cysteine (l-NAC) and rapamycin were purchased from Sigma-Aldrich.

Cytofluorimetric analysis of apoptosis

Subconfluent cells were exposed to 100 μM TN (Taurolin®; Geistlich Pharma, Wolhussen, Switzerland) or CD95-activating antibody 100 ng·mL⁻¹ (clone CH11; Upstate Biotechnology, Lake Placid, NY, USA). After 24-h incubation, cells were harvested in binding buffer (10 mM Hepes/NaOH pH 7.4, 140 mM NaCl, 2.5 mM CaCl₂), stained in the dark for 10 min with 5 μL of fluoroscein isothiocyanate-labelled annexin V (Alexis, Lausanne, Switzerland), washed with binding buffer and then stained with 1 μg·mL⁻¹ propidium iodide (Sigma-Aldrich). We analysed 5,000 events per sample. Apoptotic cells were positive to annexin V staining only and late apoptotic cells were positive to both annexin V and propidium iodide staining.

Terminal deoxynucleotidyl transferase-mediated deoxyuridine triphosphate nick-end labelling analysis

Apoptosis was evaluated by terminal deoxynucleotidyl transferase-mediated deoxyuridine triphosphate nick-end labelling (TUNEL) analysis (DeadEnd^TM Colorimetric TUNEL System; Promega, Madison, WI, USA) following treatment for 24 h with 100 μM TN alone and in the presence of 10 mM GSH or 10 mM l-NAC. In brief, subconfluent cell cultures were exposed to medium supplemented with 2% FBS according to different treatments for 24 h and fixed in 10% buffered formalin. Biotin-deoxyuridine (dU)-positive nuclei were counted on 10 fields with at least 100 cells in the same slide.

Cell cycle analysis

Cells were synchronised by 0.1 μg·mL⁻¹ colcemid (Sigma-Aldrich) treatment for 24 h, and then kept in normal medium for 4 days before analysis. After treatment with 100 μM TN for 6 h, cells were washed in phosphate-buffered saline, fixed in ethanol and stained for 30 min at room temperature with 50 μg·mL⁻¹ propidium iodide (Sigma-Aldrich) in phosphate-buffered saline containing RNase. 10,000 events per sample were analysed by flow cytometry.

Cytotoxicity and DNA adducts

Cells were treated for different times and at different drug concentrations in the presence of 2% FBS. Cytotoxicity was assessed by MTT assay, using 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (Sigma-Aldrich) and performed in quadruplicate, as previously described 17. Normalised viability percentages were obtained according to the ratio (A₅₇₀ mean values of extracts from exposed samples/A₅₇₀ mean values of extracts from control cell samples)×100. DNA adducts were evaluated by high-performance liquid chromatography on DNA extracts from cells treated with 100 μM TN for 16 h and are expressed as amount of 8-hydroxy-2-deoxyguanosine per 10⁵ deoxyguanosines, as previously described 18.

Nitrite production

Cells were cultured in medium containing 2% FBS and then stimulated with 100 μM TN in medium containing 2% FBS. Nitrite production was determined by the Griess Reagent System (Promega).

Immunoblotting

After drug treatment, subconfluent cells were lysed in L-buffer (2.5% sodium dodecyl sulfate (SDS), Tris-HCl 250 mM pH 7.4) and 40 μg of total cell lysates were loaded in reducing conditions. After separation on SDS-polyacrylamide gel electrophoresis and transfer to nitrocellulose filter (Protran; S&S, Dassel, Germany), filters were probed with phospho-Akt (Ser⁴⁷³), phospho-p70 S6 kinase (Thr³⁸⁹), phospho-p38 MAPK (Thr¹⁸⁰/Tyr¹⁸²), phospho-SAPK/c-Jun N-terminal kinase (JNK) (Thr¹⁸³/Tyr¹⁸⁵), phospho-PTEN (Ser³⁸⁰/Thr^382/383), phospho-protein phosphatase (PP)1α (Thr³²⁰), Akt, p70 S6 kinase, p38, SAPK/JNK, PTEN, PP1α, PP2A and caspase-8 antibodies (all from Cell Signaling Technology, Beverly, MA, USA), phospho-extracellular signal-regulated kinase (Erk) 1/2 (Thr¹⁸³/Tyr¹⁸⁵) and α-tubulin antibodies (both from Sigma-Aldrich), phospho-PP2A (Tyr³⁰⁷) and poly(ADP-ribose) polymerase (PARP)-1 antibodies (from Santa Cruz Biotechnology, Inc., Santa Cruz, CA, USA). The signal was detected by the enhanced SuperSignal West Pico Chemiluminescent Substrate (Pierce, Rockford, IL, USA).

Immunoprecipitation

For immunoprecipitation, after drug treatment, total cellular proteins were extracted by RIPA buffer (50 mM Tris-HCl pH 7.4, 150 mM NaCl, 0.5% sodium deoxycholate, 1% Triton X-100, 0.1% SDS) containing protease inhibitors (10 μg·mL⁻¹ aprotinin, 10 μg·mL⁻¹ leupeptin, 10 μg·mL⁻¹ pepstatin, 1 mM phenylmethylsulfonyl fluoride) and phosphatase inhibitors (1 mM Na₃VO₄, 2 mM NaF). For co-immunoprecipitation, after drug treatment, cells were lysed in solubilisation buffer (20 mM Tris-HCl pH 7.4, 5 mM EDTA, 150 mM NaCl, 10% glycerol, 1% Triton X-100) with protease and phosphatase inhibitors. 500-μg aliquots of clarified cell lysates were incubated with 1 μg of antibody immobilised on protein-A-sepharose 4B packed beads (GE Healthcare, Piscataway, NY, USA) for 2 h at 4°C. After extensive washes with lysis buffer, precipitated proteins were loaded in reducing conditions as described above. Filters were probed with Met (hepatocyte growth factor receptor) and platelet-derived growth factor receptor (PDGFR)-β antibodies (both from Santa Cruz Biotechnology), phospho-tyrosine, p85, p110α and nitro-tyrosine antibodies (all from Upstate Biotechnology), Hsp-90 (from BD Biosciences, San Jose, CA, USA).

Statistics

Data from cytotoxicity, apoptosis and cell cycle cytofluorimetric analysis, nitrite production and DNA adducts are expressed as mean±se of at least three independent experiments. Statistical differences were evaluated by ANOVA, followed by Tukey's honestly significant difference test. Values from TUNEL assay are expressed as percentages of positive nuclei over total counted. Statistical analysis was performed by Fisher's exact test. In all statistical evaluation, the significance threshold was specified in the text. All statistical tests were two-sided and calculated using Origin software (Microcal Software, Northampton, MA, USA).

TN is cytotoxic and pro-apoptotic in mesothelioma cells

Primary mesothelial (HMC), immortalised mesothelial (Met-5A) and mesothelioma cells (MMB and MMP) were treated with different concentrations of TN, ranging from 25 μM to 150 μM, for 16 h in low serum. Normal dermal fibroblasts and lung carcinoma cells were also treated as controls. A significant cytotoxic effect was observed at concentrations >50 μM in MMB, MMP and A549 neoplastic cells compared with non-neoplastic cells (p<0.001). MMB cells were more sensitive to TN, while non-neoplastic cells, whether mesothelial or not in origin, displayed negligible cytotoxicity upon the same TN treatment (fig. 1a⇓). The cytotoxic effect induced by 100 μM TN was time-dependent, starting after 5 h of treatment (p<0.001) (fig. 1b⇓).

• Download figure
• Open in new tab
• Download powerpoint

Fig. 1—

Taurolidine (TN) induces cell death in mesothelioma cells in a time- and dose-dependent manner. a) Viability (MTT) assay performed on human dermal fibroblast (•), human mesothelial cells (HMCs; ▪), MET5A (▴), malignant mesothelioma cell lines MMB (▵) and MMP (□), and A549 cells (○) treated with TN (25–150 μM), for 16 h in 2% fetal bovine serum medium. Filled symbols indicate non-neoplastic cells and empty symbols indicate neoplastic cells. The lineage of different cells are reported in the Methods section. b) Viability (MTT) assay performed at the indicated times on HMCs (▪), and MMB (▵) and MMP (□) cells treated with 100 μM TN.

TN-dependent programmed cell death was also examined. Mesothelioma cells, labelled with propidium iodide and annexin V, were analysed for apoptosis by flow cytometry in the presence of TN. Moreover, given that Fas takes part in cell death pathways in mesothelioma 19, apoptosis has also been verified following agonistic CD95 antibodies (CH-11). A significant number of apoptotic cells was observed in mesothelioma cells treated with 100 μM TN for 24 h (p<0.001). However, apoptosis induced in MMB cells was far higher than that induced in MMP cells (fig. 2a⇓). These results were confirmed by cell cycle analysis (fig. 2b⇓), which revealed a significant increase in the sub-G1 population in MMB (p<0.001) and MMP (p<0.005) cells treated with TN. Moreover, treatment with TN induced a significant decrease in G2-M subpopulation in both mesothelioma cells compared with untreated cells (p<0.001) (table 1⇓).

• Download figure
• Open in new tab
• Download powerpoint

Fig. 2—

Taurolidine (TN) is pro-apoptotic in mesothelioma cells. a) Flow cytometry analysis of apoptosis in human mesothelial cells (HMCs), malignant mesothelioma MMB and MMP cells treated with 100 ng·mL⁻¹ CD95-activating antibody (░) or with 100 μM TN for 24 h in 2% fetal bovine serum (FBS) medium (▓) after annexin V/propidium iodide (PI) labelling. □: untreated cells as compared with αCD95 (CD95 activating antibody)- and TN-treated cells. b) Representative picture of flow cytometry cell cycle analysis, performed as reported in the Methods section on HMCs, and MMB and MMP cells either treated or not treated with 100 μM TN for 6 h in 2% FBS. The percentages of cells in the different phases of the cell cycle are given in table 1⇓. c) Immunoblotting analysis of poly(ADP-ribose) polymerase (PARP) cleavage and caspase-8 activation on total cell lysates of HMCs, and MMB and MMP cells treated with 150 μM TN for 6 h in 2% FBS.

To achieve a molecular characterisation of TN-induced apoptosis, PARP and caspase-8 cleavage was determined by Western blotting on total cell lysates of HMCs, and MMB and MMP cells after 6-h treatment with 150 μM TN. Under these conditions, TN induced a clear-cut cleavage of PARP and of caspase-8 proteins in mesothelioma cells, as shown by the appearance of lower molecular weight bands (fig. 2c⇑). Apoptosis, PARP and caspase-8 cleavage or cell cycle alterations were not observed in non-neoplastic HMCs (fig. 2⇑).

We conclude that TN specifically induces apoptosis of mesothelioma cells in a time- and dose-dependent manner. The lack of effects on non-transformed mesothelial cells prompted us to verify the mechanism of TN targeting to mesothelioma cells.

TN inactivates Akt and activates PP2A in mesothelioma cells

TN has been reported to affect several intracellular pathways and to inhibit protein synthesis 7. As we and others have reported Akt as playing a key role in mesothelioma survival 4, 5, 12, we verified in mesothelioma cells the effect of TN on the activity of Akt and its downstream p70 S6 kinase (p70^S6K) effector, a known regulator of translation. TN inhibited Akt and p70^S6K phosphorylation only in mesothelioma cells (MMP and MMB), and not in non-neoplastic HMCs. On the contrary, the specific inhibitor of mTor, rapamycin, was effective in blocking the downstream p70^S6K in all cells examined (fig. 3a⇓). To verify Akt phosphorylation in cells resistant or sensitive to TN treatment, we compared non-neoplastic TN-resistant HMCs with TN-sensitive MMP mesothelioma cells displaying comparable levels of Akt protein. The inhibition of Akt activity by TN was dose-dependent, being evident at 50 μM and reaching a maximum at 150 μM, whereas in HMCs Akt phosphorylation was totally unaffected (fig. 3b⇓). The inhibitory effect of TN on Akt activity was also time-dependent, starting at 30 min and lasting for up to 8 h (fig. 3c⇓). To verify the specificity of TN signalling inhibition, we also compared the activities of Erk 1/2, JNK and p38 by immunoblotting with phosphospecific antibodies on lysates of HMCs and MMP cells treated with TN with the same kinetics evaluated for Akt. Moreover, to get a better insight into the inhibitory effect induced by TN, we examined the activities of two protein phosphatases and of one lipid phosphatase, which have been reported to regulate Akt 20. Therefore, the phosphorylation of residues critical for the activity of PP2A (Tyr307), PP1α (Thr320) and PTEN (Ser380, Thr382/383) phosphatases, all positively regulated by Ser/Thr or Tyr de-phosphorylation 21, was evaluated by immunoblotting with phosphospecific antibodies on non-neoplastic HMCs and on MMP cells. Regarding Erk 1/2 activity, beside of a sharp increase induced after 30 min of TN treatment in both cell types, no differences were observed thereafter, even upon prolonged (8 h) cell exposure to the drug. Conversely, JNK and p38 phosphorylation was increased by TN treatment both in HMCs and in MMP cells (fig. 3c⇓). Analysis of the phosphatase activities revealed that PP2A became de-phosphorylated with a slower kinetics to those of Akt, with a significant activation from the second hour of treatment, indicating a progressive activation in phosphohydrolase activity upon TN. On the contrary, PP1α showed a very late (8 h) and weak activation upon treatment with TN, whereas PTEN activity did not vary at all (fig. 3c⇓). Similar findings were found in the MMB cell line (data not shown).

• Download figure
• Open in new tab
• Download powerpoint

Fig. 3—

Taurolidine (TN) specifically inhibits the Akt pathway in mesothelioma cells. a) Immunoblotting analysis of the phosphorylation of Akt (P-Akt, Ser⁴⁷³), p70 S6 kinase (P-p70^{S6 kinase}, Thr³⁸⁹) on non-neoplastic human mesothelial cells (HMCs), on neoplastic malignant mesothelioma MMP and MMB cells treated with 200 nM of the mTOR inhibitor rapamycin (RA) or with 150 μM TN for 30 min in 2% fetal bovine serum (FBS) medium. The protein level of the single kinases and of α-tubulin are also reported as controls. b) Dose–response immunoblotting assay of P-Akt (Ser⁴⁷³) on HMCs and MMP cells treated with between 50 μM and 500 μM of TN for 30 min in 2% FBS medium. Expression of Akt and α-tubulin are also reported as controls. c) Time-course immunoblotting analysis of phosphorylation levels of Akt (Ser⁴⁷³), Erk 1/2 (Thr¹⁸³/Tyr¹⁸⁵), p38 (Thr¹⁸⁰/Tyr¹⁸²), JNK (Thr¹⁸³/Tyr¹⁸⁵), PP2A (Tyr³⁰⁷), PP1α (Thr³²⁰) and PTEN (Ser³⁸⁰/Thr^382/383) on HMCs and MMP cells treated with 150 μM TN for different times ranging from 30 min to 8 h in 2% FBS medium. The protein levels of the single effectors and of α-tubulin are also reported as loading controls.

Moreover, we tested cytotoxicity elicited by TN on MMB cells expressing a myristoylated form of Akt (myrAkt-MMB), which is permanently localised to the plasma membrane and therefore constitutively active 22. Surprisingly, no differences were observed between wild-type MMB and the same cells expressing active Akt in cell viability and in inhibiting Akt phosphorylation in a dose-dependent manner (data not shown).

Altogether, these results indicate that TN exerts a specific time- and dose-dependent inhibition on Akt activity only in mesothelioma cells, and not in normal mesothelial cells, and that PP2A jointly with PP1α might be additionally involved in sustaining this process. The increase in JNK and p38 activities is also induced by TN treatment, but it is not specific for neoplastic cells.

TN pro-apoptotic activity on mesothelioma cells is mediated by oxidative stress

TN pro-apoptotic effect on glioma cells relies upon the generation of reactive oxygen intermediates 16 and oxidative stress can cause Akt inhibition in human leukaemia cells 15. We verified whether the mechanism of TN pro-apoptotic activity in mesothelioma cells might be the activation of an oxidative pathway and whether this was responsible for TN biological effects.

The rate of oxygen free radical generation is changed by the production of nitric oxide. We evaluated nitrite production by Griess assay, upon treatment of HMCs, as well as of MMB and MMP mesothelioma cells, with 100 μM TN for different times ranging from 30 min to 24 h. We observed a time-dependent nitrite production occurring early and only in mesothelioma cells, where the differences with mesothelial cells became significant (p<0.001) at between 2 and 4 h. In HMCs, only after 24 h was a significant level of nitrite production observed, which was lower than that of mesothelioma cells (p<0.001) (fig. 4a⇓). To verify the oxidative-stress induced by TN, we measured the amount of 8-hydroxy-2′ deoxyguanosine (8-OHdG) DNA adducts, as a consequence of intracellular reactive oxygen species production 23. Upon TN treatment, the fold increase of DNA adducts, normalised for 10⁵ 2′-deoxyguanosine molecules, was significantly higher in mesothelioma cells as compared with non-neoplastic mesothelial cells (p<0.001) (fig. 4b⇓). These results confirm that mesothelioma cells are more sensitive to TN activity, which is mediated by oxidative stress.

• Download figure
• Open in new tab
• Download powerpoint

Fig. 4—

Taurolidine (TN) action on mesothelioma cells is mediated by oxidative stress. a) Griess assay for nitrite micromolar determination in human mesothelial cells (HMCs; □), and malignant mesothelioma MMP (░) and MMB (▓) cells treated with 100 μM TN at different times, ranging from 30 min to 24 h, in 2% fetal bovine serum (FBS) medium. b) 8-Hydroxy-2-deoxyguanosine (8-OHdG) adduct determination in HMCs and MMP and MMB cells treated with 100 μM TN for 16 h in 2% FBS. Data are reported as fold increase over the respective untreated sample. c) Immunoblotting analysis of phosphorylation of Akt (Ser⁴⁷³) in MMP cells in 2% FBS medium pre-treated for 24 h with 10 mM glutathione mono-ethylester (GSH) or with 10 mM l-N-acetyl-cysteine (l-NAC), respectively, and exposed to 150 μM TN for 30 min and to both TN and antioxidants (upper panel, TN+GSH and lower panel, TN+l-NAC). Levels of Akt and α-tubulin proteins are reported as loading controls. d) Dose–response viability (MTT) assay, performed on MMP cells in 2% FBS medium with TN ranging from 50 to 150 μM for 24 h or treated with TN at the same concentrations in presence of 10 mM GSH or l-NAC. e) terminal deoxynucleotidyl transferase-mediated deoxyuridine triphosphate nick-end labelling (TUNEL) assay performed on MMP cells in 2% FBS medium, with 100 μM TN for 24 h, in either the presence or absence of GSH 10 mM or l-NAC 10 mM. Percentages of apoptotic nuclei are shown, following counts on 10 fields with at least 100 cells on the same slide.

Antioxidant agents have been widely used to prevent the effects elicited by oxidative stress in live cells 24. We evaluated Akt phosphorylation in lysates from MMP cells pre-treated with 10 mM GSH for 24 h before the addition of 150 μM TN for 30 min. GSH did not affect Akt phosphorylation in the absence of TN, whereas GSH cell pre-treatment completely prevented Akt inhibition by TN, suggesting that Akt dephosphorylation by TN is mediated via an oxidative response (fig. 4c⇑). Similar results were obtained in a parallel experiment conducted using the antioxidant agent l-NAC (fig. 4c⇑). These antioxidant agents did not alter TN-induced JNK and p38 activation (data not shown).

Cell viability was examined by MTT assay in cells treated for 24 h with increasing concentrations of TN, after 24 h of pre-treatment with 10 mM GSH or 10 mM l-NAC. Under these conditions, TN induced a dose-dependent decrease in cell viability, which is more pronounced than that observed after only 16 h (see fig. 1a⇑). Pre-exposure of the same cells to the antioxidant agents significantly inhibited the cytotoxic effect of TN, even at the highest concentration (p<0.001) (fig. 4d⇑). The effect of antioxidants on apoptosis was also evaluated using TUNEL assay on MMP cells. A higher number (14.1%) of biotin-dU-positive nuclei was observed upon cell exposure to 100 μM TN for 24 h than in control nonstimulated cells (3.1%). The concurrent treatment with 10 mM GSH or 10 mM l-NAC decreased the percentage of apoptotic nuclei to 2.6% and 3.1%, respectively (fig. 4e⇑). These differences were statistically significant (p<0.001). We obtained similar results for cell viability and apoptosis, with both antioxidant agents acting on the mesothelioma MMB cells (data not shown).

We conclude that the cytotoxic, pro-apoptotic effect of TN on mesothelioma cells, stems from oxidative stress, involves nitrite production and is reversed by general antioxidant agents.